This is an archived low-energy page for bots and other anonmyous visitors.
Please sign up if you are a human and want to interact.
in reply to Reliable software: SOLVED (was: Reliable software OR Is CPAN the sacred cow)
I am much obliged to all monks here. Now I've answer which I looking for in my initial post!
My initial post was very incorrectly formulated, but that's because to formulate question correctly you need to know at least half of answer, which I did't know at that time.
My initial post (and most comments) was deliberately designed as rant, because I've feeling I need much more than usual 3-5 comments to find answer. I'm sorry about this, but it's worked! :)
Now, in hope this helps somebody in the future, I'll repeat my question (better formulated and without rant), with answer.
Q: Why not develop everything with maximum possible reliability and security? Isn't world become better if our OS will be designed like Inferno and all software will be as reliable and secure as DJB's?
A: Because there different goals in development:
- If you need to "have your work done", then "works for me" is right way.
This way your software will be as reliable and secure as spec require, no more and no less. If you wish to improve overall software quality in the world in this case, you must try to increase spec requirements to reliability and security (i.e. go talk with your boss/customer).
- If you need to get maximum possible reliability&security, like DJB do, then you need to have a lot of time, experience, and be able to change spec if it require something you can't develop as reliable&secure as you wish.
This is more like intellectual game.
Such sort of development doesn't even guaratee you'll "have your work done" at some point, a good example of this is unfinished DJB's Internet Mail 2000 project. Also it doesn't guarantee your software will be useful for somebody, because people may need features which you've removed from spec (and that's why we've so many custom patches for Qmail).
There also exists slightly more abstract answer. If you experienced enough, then you able to develop software per ANY spec. If spec will require DJB-like quality - no prob. If spec will require low quality cheap software developed very quickly - no prob. This way we've no more different goals, just different specs. And it's always up to you, agree to THAT spec, or not, and look for different work instead. |
Re^2: Reliable software
by Asim (Hermit) on Sep 16, 2006 at 18:10 UTC
|
My initial post (and most comments) was deliberately designed as rant, because I've feeling I need much more than usual 3-5 comments to find answer. I'm sorry about this, but it's worked! :)
I don't know about the rest of the Monks here, but for myself? I respect those who respect me. And ranting on here, during my time, to "get a rise"? To get a "full answer". That's not respecting the other people on here. Especially as I take time from my actual job to answer questions on here, many times.
If you had asked for help in a respectful manner, you would have received a ton of replies that would have pointed you to source after source. If you had chosen the path of asking honest questions, rather than posturing, you would have received honest and detailed answers that would have not only enlightened you, but saved you time and energy.
I'm going to save my reply to your questions to me for another day, perhaps. I'll close this by passing on something I keep in my Personal Nodelet, Spirit of the Monastery. I can only recommend a reading, and pondering, of it's words, as I do, from time to time, before posting.
----Asim, known to some as Woodrow.
| [reply] |
Re^2: Reliable software
by badaiaqrandista (Pilgrim) on Sep 18, 2006 at 06:37 UTC
|
My initial post was very incorrectly formulated, but that's because to formulate question correctly you need to know at least half of answer, which I did't know at that time.
You've answered your own question why people put unreliable and not-secured code in CPAN: because to formulate something correctly you need to know at least half of answer, which you did not know at that time. People put code in CPAN for different reason and different purpose and different contexts. Therefore their priorities are different, while in your case, it's reliability and security.
My initial post (and most comments) was deliberately designed as rant, because I've feeling I need much more than usual 3-5 comments to find answer. I'm sorry about this, but it's worked! :)
You've just made a very bad reputation of yourself by doing this. There are a lot nicer ways of getting a lot of comments from fellow monks than what you've done here. Social skill is a lot more important than programming skill when you are in a community, online or not.
| [reply] |
|
|
There are a lot nicer ways of getting a lot of comments from fellow monks than what you've done here.
You know this, or this is just your hope?
Top nodes from last 300 Meditations ordered by number of 1st level comments:
33/161 Perl is dying
23/59 this node
21/70 O'reilly some sort of perl monopoly?
20/61 The Perl Hacker Inferiority Complex
20/54 The REAL reason for why they choose PHP over Perl.
19/41 Let's face it, Perl *is* a scripting language
16/69 Perl in the Enterprise
13/56 Secret Perl Operators: the boolean list squash operator, x!!
... and ordered by overall number of comments:
33/161 Perl is dying
21/70 O'reilly some sort of perl monopoly?
16/69 Perl in the Enterprise
20/61 The Perl Hacker Inferiority Complex
23/59 this node
13/56 Secret Perl Operators: the boolean list squash operator, x!!
20/54 The REAL reason for why they choose PHP over Perl.
19/41 Let's face it, Perl *is* a scripting language
So, you see, there is only one node (Secret Perl Operators: the boolean list squash operator, x!!) which isn't use ranting in any way, and it isn't in 'top 5' in both lists. I don't try to say been in 'top 5' "is goodness", :) I try to say that node is unique enough and sort of singularity.
You've just made a very bad reputation of yourself by doing this.
BTW, guess who is author of top nodes? Right. Anonymous Monk. That's probably because their authors worry about their reputation on PM, while I'm just worrying about development reliability, security and overall quality, even if this costs me some reputation points.
2006-09-20 Retitled by planetscape, as per Monastery guidelines
Original title: 'ranting'
| [reply] |
|
|
By bad reputation he didn't mean your rep points, he meant your actual reputation with the monks here. How good of response do you think you'll get next time you have a problem? BTW Number of nodes isn't realy a usefull way to tell if something was answered. You could have phrased this right and gotten a single response answering it. Instead you now have a bunch of monks who just wont bother to try and help you in the future.
| [reply] |
|
|
|
|
|
|
|
|
It seems that you've confused quality and quantity. They sure got a lot of comments, but can you learn much from them?
Your "solution" is a good way to ask. And that's the kind of answer you'd get. You wont get a lot of answer, but you can learn from that. Isn't that better than pissing people off just to get a simple answer like that?
| [reply] |
|
|
